How to high current solid state(mosfet) relay

Thread Starter

bigjoncoop

Joined Feb 1, 2019
129
Hey guys,

Looking for a circuit presumably using a bunch of high current mosfets in order to create a " solid state relay" rated about 300a...

Trying to make a relay for my DIY spot welder. There are a bunch of circuits I found on Google but most of them are also implementing a timer circuit as well.

I just need to be able to have it trigger when 12v is applied. Instantly turn on when 12v is present and instantly turn off with 12v is not.

I understand I should probably use a bunch of fast switching power mosfets in series. Just not sure if I should be using p Channel or n Channel. And any other characteristics I should be looking for. Also something that is fairly common so it's easy to source and low cost

Any advice is greatly appreciated

Thank you
 

Papabravo

Joined Feb 24, 2006
14,393
Problem is that "instantly" and MOSFET do not go together well. A MOSFET gate looks like a big capacitor, and a big capacitor needs time to charge and discharge. Fortunately the datasheet (remember those) has the information you need to estimate these times. While all this is going on there is a huge amount of power being dissipated in the MOSFET(s). You need to replace your current vocabulary with something a bit more realistic. Maybe you should spend some time on a simulator to see just how fast you can make a MOSFET switch. then at least your request would have some credibility.
 

sghioto

Joined Dec 31, 2017
1,443
Any advice is greatly appreciated
As suggested in your previous thread I would look at the IXTK200N10L2 mosfet. Two of these paralleled with proper drivers.
Also previously suggested, I would strongly advise controlling the mosfets using a timer circuit.
SG
 

Papabravo

Joined Feb 24, 2006
14,393
As suggested in your previous thread I would look at the IXTK200N10L2 mosfet. Two of these paralleled with proper drivers.
Also previously suggested, I would strongly advise controlling the mosfets using a timer circuit.
SG
That being the case, the TS should, as the man in black said: "Get used to disappointment".
 

MaxHeadRoom

Joined Jul 18, 2013
21,402
Traditional/Industrial spot welders use a mains transformer and control the primary with a Triac or other device etc,
The secondary is only one or two turns and a few hundred amps, for E.G. Your Weller solder gun has virtually the same construction, the secondary is one turn and can easily reach 100amps.
A micro wave transformer has been used in the past by DIY'ers.
There is also the capacitor DC discharge method that uses a SCR.
Max.
 

Papabravo

Joined Feb 24, 2006
14,393
What does that mean?
SG
It is a movie cliché. (Princess Bride, 1987) It refers to the problem with a qualitative term such as instantaneous. If the TS cannot be more precise, then he is going to be "disappointed" in a circuit that switches from on to off and off to on in milliseconds. For some MOSFETS this is what you get, and there is no practical way around it. Understanding the behavior and careful reading of a datasheet can mitigate, but not eliminate this problem. That is what it means.
 

Thread Starter

bigjoncoop

Joined Feb 1, 2019
129
Wow thanks everybody for your input. I guess I should have been a little clearer on my initial question.

I tried to answer your questions below. and I have came up with a mosfet circuit that is proven to work in the DIY Community. Using the STP110N7F6 mosfet. It has been thoroughly tested. I've been doing some research and it seems like the most common reason these type of mosfet circuits could fail is from Avalanche current? When the mosfet is abruptly shut off with such high current flow, it needs a way to escape basically it's my understanding. It seems normally a schottky diode would be placed in parallel with the gate resistor.

now this mosfet circuit has been used just the way it is with no diode at all and again has been working well for many people. I would rather spend a few extra dollars to protect all the mosfets if it's as simple as adding a diode or a few of them.

I'm having a hard time sizing the shockey diode for this circuit. and I have seen some examples with the diode in parallel with the gate resistor like mentioned above, and It seems normally a schottky diode would be placed in parallel with the gate resistor, but I've also seen examples with a diode on each mosfet. not sure which way is correct.

also in the data sheet for this particular mosfet shows a gate resistor value of 4.7ohm. but in the details from the DIY community they say to use a 2.2ohm. and somewhere I remember seeing for this particular mosfet it said to use it gate resistor of 100ohm to 1kohm but can't find where I read that.

I've included a picture of the schematic and the mosfet ratings and characteristics as well as a link to the datasheet. normally people are using 12 of these mosfets in parallel. I've only included a few in my schematic just to show the general layout. I understand that using 12 of them is super Overkill most likely. the battery I am using is only rated at 250cca. in your opinion what would be the amount of mosfets needed to handle that amount of current including some overhead? Or should I just stick with using 12 of them

was about to place an order with digikey, but wanted to figure out these few unknowns before I do so.
thanks again for everything.

Capture.JPG
Mos1.JPGMos2.JPGMos3.JPG

Problem is that "instantly" and MOSFET do not go together well.
I was trying to get at by saying instantly was, fairly quickly not a delay of a few seconds. Basically a fast switching mosfet

As suggested in your previous thread I would look at the IXTK200N10L2 mosfet. Two of these paralleled with proper drivers.
Also previously suggested, I would strongly advise controlling the mosfets using a timer circuit.
SG
I can get way with only using two mosfets? I guess why not if they're rated High Enough. And I do have a timer circuit already. If one of those timer boards with a relay. I have it set to close the contacts for so many milliseconds. Works fine. The Solenoid I was using seems to have burned up its contacts. That's why I'm thinking of going the mosfet route.

Traditional/Industrial spot welders use a mains transformer and control the primary with a Triac or other device etc,
The secondary is only one or two turns and a few hundred amps, for E.G. Your Weller solder gun has virtually the same construction, the secondary is one turn and can easily reach 100amps.
A micro wave transformer has been used in the past by DIY'ers.
There is also the capacitor DC discharge method that uses a SCR.
Max.
You are correct. During my initial research of a DIY battery spot welder there are many tutorials on using a microwave Transformer. But for various reasons I feel that is definitely not the safest and most logical route for myself to take for numerous reasons.
 
Last edited:

sghioto

Joined Dec 31, 2017
1,443
I can get way with only using two mosfets? I guess why not if they're rated High Enough
The IXTK200N10L2 are rated at 200 amps but expensive. I would stay with the STP110N7F6 they are less than a dollar apiece.
SG
EDIT: Your schematic is incorrect. Should be as below.
EEE DIY 18650 spot welder.png
 
Last edited:

Thread Starter

bigjoncoop

Joined Feb 1, 2019
129
The IXTK200N10L2 are rated at 200 amps but expensive. I would stay with the STP110N7F6 they are less than a dollar apiece.
SG
EDIT: Your schematic is incorrect. Should be as below.
View attachment 215944
Yes thank you. I know better. Didn't notice.

Do I need to use 12 of them though for my 250a battery?(power source)

And any ideas on protection for the mosfets ?
 

sghioto

Joined Dec 31, 2017
1,443
Do I need to use 12 of them though for my 250a battery?(power source)
And any ideas on protection for the mosfets ?
I would think 6 should be enough but again less than a dollar each.
I think you're fine with just the resistors. It's not high speed switching.
SG
 
Last edited:

Papabravo

Joined Feb 24, 2006
14,393
The IXTK200N10L2 are rated at 200 amps but expensive. I would stay with the STP110N7F6 they are less than a dollar apiece.
SG
EDIT: Your schematic is incorrect. Should be as below.
View attachment 215944
You have the input going to the Source of the left most device and to the GATE of all the others. It might work but not the way you want it to. If you can't be more careful with schematics I doubt very much you will get anything of this magnitude to work. that is if you don't injure or kill yourself first.
 
Top